Multiprocessors

1
Multiprocessors

• Andreas Klappenecker
• CPSC321 Computer Architecture

2
Idea

• Build powerful computers by connecting
• many smaller ones.

3
Multiprocessors
Good for timesharing easy to realize -
difficult to write good concurrent programs -
hard to parallelize tasks - mapping to
architecture can be difficult
4
Questions

• How do parallel processors share data? single
  address space message passing
• How do parallel processors coordinate?
  synchronization (locks, semaphores) built into
  send / receive primitives operating system
  protocols
• How are they implemented? connected by a
  single bus connected by a network

5
Shared Memory Multiprocessors
Problems???
Symmetric multiprocessor (SMP)
6
Distributed Memory Multiprocessors

• Distributed shared-memory multiprocessor
• Message passing multiprocessor

7
Multiprocessors
8
Connection Network

• Static Network
• fixed connections between nodes
• Dynamic Network
• packet switching (packets routed from sender to
  recipient)
• circuit switching (connection between nodes can
  be established by crossbar or switching network)

9
Static Connection Networks
10
Circuit Switching Delta Networks
0101

• Route from any input x to output y by selecting
  links determined by successive d-ary digits of
  ys label.
• This process is reversible we can route from
  output y back to x by following the links
  determined by successive digits of xs label.
• This self-routing property allows for simple
  hardware-based routing of cells.

1
1
0
1
0
1
1101
0
1
xxk-1 . . . x0
yyk-1 . . . y0
11
Network versus Bus
12
Performance / Unit Cost
13
Programming

• lock variables
• semaphores
• monitor

14
Cache Coherency
15
Outlook

• Distributed Algorithms
• Distributed Systems
• Parallel Programming
• Parallel Compilers